Enhanced activation of peroxymonosulfate by Fe/N co-doped ordered mesoporous carbon with dual active sites for efficient removal of m-cresol

Donghui Li,Wenzhe Wu,Xue Ren,Xixi Zhao,Hongbing Song,Meng Xiao,Quanhong Zhu,Hengjun Gai,Tingting Huang,*

1 College of Chemical Engineering,Qingdao University of Science and Technology,Qingdao 266042,China

2 School of Chemical &Environmental Engineering,China University of Mining &Technology-Beijing,Beijing 100083,China

Keywords: Degradation Peroxymonosulfate Fe(II)/Fe(III)/FeN4 Ordered mesopores carbon Catalyst Radical

ABSTRACT The novel Fe-N co-doped ordered mesoporous carbon with high catalytic activity in m-cresol removal was prepared by urea-assisted impregnation and simple pyrolysis method.During the preparation of the Fe-NC catalyst,the complexation of N elements in urea could anchor Fe,and the formation of C3N4 during urea pyrolysis could also prevent migration and aggregation of Fe species,which jointly improve the dispersion and stability of Fe.The FeN4 sites and highly dispersed Fe nanoparticles synergistically trigger the dual-site peroxymonosulfate(PMS)activation for highly efficient m-cresol degradation,while the ordered mesoporous structure of the catalyst could improve the mass transfer rate of the catalytic process,which together promote catalytic degradation of m-cresol by PMS activation.Reactive oxygen species(ROS)analytic experiments demonstrate that the system degrades m-cresol by free radical pathway mainly based on and ?OH,and partially based on ?OH as the active components,and a possible PMS activation mechanism by 5Fe-50 for m-cresol degradation was proposed.This study can provide theoretical guidance for the preparation of efficient and stable catalysts for the degradation of organic pollutants by activated PMS.

1.Introduction

With the vigorous development of the coal chemical industry,the problem of water pollution continues to increase [1].m-Cresol is one of the main pollutants in coal chemical wastewater,which has the characteristics of high toxicity and refractory degradation,the effective removal ofm-cresol is of great significance to the ecological environment [2,3].

In recent years,many advanced oxidation methods have emerged in wastewater treatment,such as photocatalytic oxidation [4,5],Fenton oxidation [6–8],supercritical catalytic oxidation[9],catalytic wet oxidation[10,11]and sulfate radical()based advanced oxidation process (SR-AOP) [12–16].Among them,although photocatalytic oxidation can effectively degrade phenol,the unstable activity of photocatalytic materials hinders their reuse.Moreover,as a homogeneous free radical reaction,Fenton oxidation has a significant effect on reducing the chemical oxygen demand(COD)in wastewater,however,due to the production of a large number of iron containing sludge or sewage,the promotion of this technology is greatly limited in the practical application[6].And supercritical catalytic oxidation needs to be operated under high temperature and high pressure,and the reaction dynamics and mechanism of supercritical chemical have not been fully studied[9,17].In addition,catalytic wet air oxidation technology has low mineralization ability for phenolic compounds and is prone to produce more toxic products [18].Compared with the above-mentioned advanced oxidation technologies,the SR-AOP technology has become a research hotspot for the oxidative degradation of phenolic organics due to the advantages of high treatment efficiency,mild operating conditions and simple equipment[15,19,20].Generally,SR-AOP can activate persulfate(peroxydisulfate (PDS) and peroxymonosulfate (PMS)) by means of additional energy,homogeneous catalysis,heterogeneous catalysis [21,22],etc.,to produce strongly oxidizing,so as to oxidatively degrade organic pollutants in water.Since the redox potential of(E0=2.5–3.1 eV)is higher than that of ?OH(2.8 eV),the degradation ability of former is stronger;and the half-life period of(4 s)is longer than that of ?OH(1 μs)[20],so it is easier to contact and react with organic pollutants and has outstanding treatment efficiency in the oxidative degradation of phenolic pollutants.

The transition metal-based heterogeneous catalyst activation strategy is the most common and effective activation method in SR-AOP technology [23,24].At present,a large number of transition metal based catalysts (such as Fe3O4and Fe2O3) have been developed for PMS activation because of their low cost and low toxicity,but they are often accompanied by severe metal leaching,slow catalytic kinetics and low atomic utilization,which obviously hinder their further industrial application [25].Therefore,there is an urgent requirement to develop catalysts with good catalytic activity and stability.In the preparation process,the metalsupport interaction can not only ensure the metal atoms obtain unique electronic structure,but also effectively prevent the metal atoms from forming large nanoparticles,so as to maximize the atomic efficiency.At present,the most widely used support include graphene,metal–organic frameworks(MOFs) [26,27],mesoporous silica [28] and porous carbon [29].Among them,ordered mesoporous carbon(OMC) materials are the preferred support to effectively improve atom utilization due to their large specific surface area,adjustable pore size and excellent mass transfer performance[30,31].In addition,heteroatom doping (such as N,S,P or B) can adjust the surface properties of carbon support and break the electrical neutrality of sp2hybrid carbon,which becomes an effective way to construct carbon-based catalysts with high activity and stability [32–34].Gaoet al.[35] researched N atom doped carbonsupported catalysts decorated with different transition metal centers (Mn,Fe,Co,Ni,Cu) and found that catalysts with Fe as the active metal center showed the best performance,the N atoms in the catalyst can not only improve the conductance of the catalyst,but also improve the leaching resistance of the active metal center.However,due to the complex structure and composition of Fe-NC catalysts and insufficient understanding of active sites,the synergistic mechanism of Fe and N species adsorbing organic pollutants and activating PMS is still unclear,which greatly hinders the rational design and preparation of the Fe-NC catalysts with high activity and stability.

Based on the above considerations,a Fe-N co-doped carbon material with ordered mesoporous channels prepared by ureaassisted impregnation and simple pyrolysis method was designed.During the catalyst preparation,the complexation of N elements in urea could anchor Fe,and the formation of C3N4during urea pyrolysis could prevent migration and aggregation of Fe species,which jointly improve the dispersion and stability of Fe.The synergistic catalytic effect of FeN4sites and highly dispersed Fe nanoparticles realized the dual-site PMS activation for highly efficientm-cresol degradation,while the ordered mesoporous structure of the catalyst could improve the mass transfer rate of the catalytic process,which together promoted PMS activation performance for the degradation ofm-cresol.Through electron paramagnetic resonance(EPR) and quenching experiments,it was demonstrated that the system degradesm-cresol by free radical pathway withand ?OH as the main active components and a small amount ofin the system plays an auxiliary role,and a possible PMS activation mechanism by 5Fe-50 form-cresol degradation was proposed.

2.Experimental

2.1.Material

Material details are displayed in Supplemental Material.

2.2.Preparation of catalysts

2.2.1.Preparation of surface-functionalized ordered mesoporous carbon CMK-3-N

To begin,2.0 g of P123 was added to 65.0 g of 2 mol?L-1HCl and agitated for 4 h at 40°C,followed by the addition of 4.28 g of tetraethyl orthosilicate (TEOS) to the aforesaid mixture drop by drop.After that,the mixture was agitated for 24 h at 40 °C.It was then moved to an autoclave with a 100 ml polytetrafluoroethylene(PTFE) liner and placed in a 100 °C oven for 24 h.After that,the product was filtered,rinsed in deionized water,and dried at 120 °C overnight.And the dried product was roasted in a muffle furnace at 550°C for 6.0 h to produce the ordered mesoporous silica SBA-15.

2.0 g of SBA-15 was added to a mixture of 2.5 g of sucrose,0.28 g of H2SO4and 10.0 g of H2O.The combination was dried for 6 h at 100 °C,then heated to 160 °C for another 6 h to partly polymerize and carbonize the sucrose in the silica sample.The mixture was added to a solution of 1.6 g of sucrose,0.18 g of H2SO4and 10.0 g of H2O and heated with the carbonization and pyrolysis finished by heating to 900°C under vacuum as indicated above.To remove the SBA-15 template,the carbon/silica composite product was washed twice with 5.0% (mass) HF at room temperature.The template-free carbon product was filtered,washed three times with an ethanol–water solution,and completely dried at 100 °C to get the ordered mesoporous carbon CMK-3.

0.6 g of CMK-3 was added to 30.0 ml of 3.0 mol?L-1HNO3solution,agitated and mixed well,and then the suspension was stirred in a water bath at 80 °C for 4 h;then the sample was cooled,filtered,washed with deionized water to neutrality,and dried under vacuum at 110 °C for 12 h to obtain the surface functionalized ordered mesoporous carbon CMK-3,named CMK-3-N.

2.2.2.Preparation of Fe and N co-doped ordered mesoporous catalysts with different N doping amounts

First of all,0.10 g of CMK-3-N,a certain quantity of urea(0,25.0,50.0 and 75.0 mg) and 1.50 ml of FeCl3solution (9.66 mg?ml-1)were dispersed in 20.0 ml of deionized water and stirred at 25 °C for 12 h.Then,the sample was filtrated,washed with deionized water and dried under vacuum at 80°C for 2 h.Finally,the sample was calcinated at 550 °C for 4 h and the temperature raising rate was 1 °C?min-1.The obtained product is denoted 5Fe-x(where 5 represents the amount of iron added during the preparation process,mg;xis the mass of urea,mg).

For comparison,the non-Fe doped 0Fe-50 catalyst was produced by adding 50 mg of urea to the precursor solution,other conditions remained unchanged as described above.

2.3.Characterization

Nitrogen adsorption and desorption isotherms of catalysts were measured using the Autosorb-iQ instrument.Raman spectroscopy was performed on a Horiba Jobin-Yvon LabRAM HR800 confocal Raman microscope with a laser of λ=532 nm.X-ray diffraction(XRD)analysis was recorded on an Ultima IV X-ray polycrystalline powder diffractometer in the ranges of 2θ=5° –80° and 2θ=0° –5°for wide-angle (WAXRD) and small-angle XRD analyses (SAXRD),respectively.The morphology of the sample was determined using a FEI Quanta 400 FEG (FEI,USA) scanning electron microscope(SEM).The microstructure of the catalysts was observed using a FEI Tecnai G2 F20 high resolution transmission electron microscope (HRTEM).X-ray photoelectron spectroscopy (XPS) was performed on a Thermo Fischer Scientific Escalab 250 XI equipment with monochromatic Al Kα radiation(1486.6 eV)source.The characterization details are displayed in Supplementary Material.

2.4.Catalyst performance evaluation

In a 100 ml beaker,catalytic oxidative degradation ofm-cresol was performed.To initiate the reaction,20 mg of catalyst was added to the mixture containing PMS andm-cresol,followed by continuous magnetic stirring until the reaction was complete.During the reaction,5 ml of samples were obtained at predetermined intervals and the reaction was stopped by 0.5 mL of methanol.A 0.22 μm PTFE membrane was used to filter the samples,and 1 ml of the filtrate was collected for analysis.

The pH of the solution was adjusted by 3.0 mol?L-1H2SO4or 3.0 mol?L-1NaOH to investigate the influence of pH on the catalytic performance form-cresol degradation.In addition,with the approach described above,the effects of catalyst content,nitrogen doping amount,initialm-cresol concentration,PMS concentration,initial pH of the reaction,and the presence of inorganic anions onm-cresol degradation were also examined.

Them-cresol concentration was measured at 510 nm by 4-aminoantipyrine spectrophotometry with UV–Vis spectrophotometer: phenolic compounds react with 4-aminoantipyrine in the presence of potassium ferricyanide in a medium (pH=10)and produce an orange-red substance with maximum absorbance at 510 nm.The conversion rate ofm-cresol (Xmc) was calculated as follows:

3.Results and Discussion

3.1.Characterization of catalysts

3.1.1.WAXRD and SAXRD

The wide-angle X-ray diffraction patterns of the 5Fe-0,5Fe-25,5Fe-50 and 5Fe-75 catalysts are shown in Fig.1(a).For all samples,two broad diffraction peaks are found near 2θ=24°and 44°,which are correspond to the(0 0 2) and (1 0 1) crystal planes of graphite carbon [36].And their wide half-peak widths indicate the low graphitization degree of CMK-3-N.Furthermore,there exist the characteristic peaks attributed to Fe3O4at 2θ=35.5° in 5Fe-0 and 5Fe-25 (JCPDS No.85-1436),indicating that they have large particle sizes and low dispersions of Fe3O4[37,38].In contrast,the characteristic peak of Fe3O4is not detected in the XRD patterns of 5Fe-50 and 5Fe-75,demonstrating that these catalysts have small particle sizes and high dispersions of Fe3O4.In summary,it can be concluded that as the N doping amount increases,the 5Fe-xcatalyst has smaller Fe3O4particles and better Fe3O4dispersion.

Fig.1.(a) Wide-and (b) small-angle XRD patterns of the samples.

The small-angle X-ray diffraction patterns(SAXRD)of the 5Fe-0,5Fe-25,5Fe-50 and 5Fe-75 catalysts are shown in Fig.1(b).All samples have a diffraction peak near 2θ=1°,indicating that all samples have a hexagonal symmetric ordered mesoporous structure[39],and the doping of Fe and N does not destroy the structure of CMK-3-N.However,compared with CMK-3-N(Fig.S1 in Supplementary Material),the diffraction peak intensities of the catalysts are significantly reduced,and the width of the diffraction peaks is also increased,which may be because in the preparation process,the doped iron species and C3N4generated by urea decomposition are attached to the pore wall of CMK-3-N,leading to the roughness of the pore surface or the collapse of part of the pore structure,consequently,the structural order of CMK-3-N is reduced [40,41].In addition,with the increasing of urea dosage,the small angle diffraction peak intensity of the catalysts slightly decrease,which indicates that the more C3N4content embedded in CMK-3-N support channels,the influence on the order of CMK-3-N is greater[42].

3.1.2.Raman

To obtain the degree of graphitization of carbon in catalysts,Raman spectroscopy was adopted to analyze the structural information [43].The Raman spectra of 5Fe-0,5Fe-25,5Fe-50 and 5Fe-75 are shown in Fig.2.It can be seen from the figure that all samples exhibit two characteristic peaks at 1340 and 1598 cm-1,corresponding to D band (defective carbon) and G band (graphitic carbon),respectively [44].Generally,the intensity ratio of D band to G band (ID/IG) is used to judge the disorder degree and defects of carbon materials,the largerID/IGis,the higher defect degree of carbon materials possesses[44].And the defects of carbon materials have been widely confirmed to play an important role in PMS activation.To distinguish whether the defect is caused by the doping of metal elements or the annealing temperature,the Raman characterization of non-Fe doped catalysts (0Fe-50) was also added.It can be observed that the 0Fe-50 catalyst also exhibits characteristic peaks corresponding to the D and G bands at 1340 and 1598 cm-1(Fig.S3),which indicates that the carbon defects can be obtained by annealing treatment without doping with Fe element.Subsequently,theID/IGvalue is calculated to be 0.796 and found to be smaller than theID/IGvalue of 5Fe-50 (0.823),which proves that the doping of Fe element can also introduce carbon defect structures [45–47].By integrating the characteristic peak area of the 5Fe-xcatalyst,it is calculated that theID/IGratios of 5Fe-0,5Fe-25,5Fe-50 and 5Fe-75 are 0.742,0.747,0.823 and 0.879,respectively.The reason for this phenomenon is because with the increasing urea addition,the amount of C3N4attached on the pore surface and the doped N atom gradually increase,which promote the defect degree of the carbon material and lead to the increase ofID/IGratio [48,49].

Fig.2.Raman spectra of the 5Fe-x catalysts.

3.1.3.Low temperature N2 adsorption and desorption

Fig.3 shows the N2adsorption–desorption isotherms and pore size distributions of the 5Fe-xcatalysts with different N doping amounts.The N2adsorption–desorption isotherms of 5Fe-x(Fig.3(a))are all type IV isotherms,and there are obvious H4 type hysteresis loops under medium and high relative pressures,which is similar to the CMK-3-N support,indicating the coexistence of micropores and mesopores in 5Fe-x[50].It also confirms that the doping of Fe and N does not affect the framework of the support.Table 1 shows the pore parameters of the 5Fe-xcatalysts calculated according to the Brunauer–Emmett–Teller (BET) equation and the Barret–Joyner–Helenda(BJH)method.The pore volume of the catalysts is between 1.01 and 1.06 cm3?g-1,among which the pore volume of 5Fe-0 is the largest(1.06 cm3?g-1),and with the increasing N doping,the pore volume of the sample gradually decreases.In addition,the specific surface area of the catalyst is an important parameter for studying the mass transfer of the substrate in the catalytic system and increasing the number of active sites [51].It can be seen from Table 1 that the specific surface area of the 5Fe-xcatalysts ranges from 1099 to 1156 m2?g-1.Similarly,with the increasing N doping,the specific surface area of the catalysts also shows a downward trend,which may be attributed to the increasing defect degree on the pore structure of CMK-3-N.However,compared with 5Fe-0,the decrease of pore volume and specific surface area of the 5Fe-25,5Fe-50 and 5Fe-75 catalysts is limited,indicating that Fe and N doping can only modify the pore surface of CMK-3-N without affecting the main structure of carbon material.The porous structure in the 5Fe-xcatalyst can not only provide rich channels to promote the mass transfer process of reactive substances,but also expose the active sites inside the catalyst to the maximum extent,and make full use of the nanodomain limiting effect to enhance the adsorption,collision and reaction process of reactant molecules on the catalyst.

Table 1 Textural parameters of samples with different N content

Fig.3.(a) N2 adsorption/desorption isotherms and (b) corresponding pore diameter distribution curves of 5Fe-0,5Fe-25,5Fe-50 and 5Fe-75.

3.1.4.TEM and SEM

The TEM images of the 5Fe-0 and 5Fe-50 catalysts are shown in Fig.4.It can be clearly observed that the 5Fe-0 and 5Fe-50 catalysts all show well-ordered mesoporous structure,indicating that the doping of iron and nitrogen does not seriously change the order of the carbon support,which is benefit to improve the mass transfer during the catalytic experiment.Comparing the TEM images of 5Fe-0 and 5Fe-50,it is found that there exist agglomerated larger Fe nanoparticles on the surface of 5Fe-0,while the Fe nanoparticles supported on the surface of 5Fe-50 have no agglomeration and are highly dispersed.The smaller Fe nanoparticles on the surface of 5Fe-50 may be due to the addition of urea in the synthesis process of 5Fe-50,and the N element has an anchoring effect on Fe,so that Fe atoms enter the carbon skeleton and form a FeN4complex with the doped N atoms in the form of a single atom.

Fig.4.TEM images of (a) 5Fe-50 and (b) 5Fe-0.

According to the SEM images of 5Fe-0 and 5Fe-50 (Fig.5(a)and (b)),the catalysts all maintain a rope-like shape forming long string-like aggregates [39],which is similar to CMK-3-N (Fig.S2),indicating that the doping of iron and nitrogen does not damage the basic structure of CMK-3-N and there is no Fe aggregation on the surface of 5Fe-50 after N doping.In addition,the element mapping images of Fe,N,C and O of 5Fe-50 (Fig.5(c)–(f)) show that the Fe and N elements are uniformly dispersed in the catalyst.

Fig.5.SEM images of (a) 5Fe-0 and (b) 5Fe-50.(c)–(f) Elemental mapping (Fe,N,C,O) of 5Fe-50.

3.1.5.XPS

XPS analysis is performed to further explore the elemental composition of 5Fe-0,5Fe-25,5Fe-50 and 5Fe-75 samples.The highresolution XPS spectra of Fe 2p is shown in Fig.6.5Fe-0,5Fe-25,5Fe-50 and 5Fe-75 all show six characteristic diffraction peaks,which can be convolved into three types of peaks.Among them,a pair of characteristic diffraction peaks centered at 710.1 (Fe 2p3/2) and 723.7 eV (Fe 2p1/2) belong to Fe2+,and a pair centered at 712.9 (Fe 2p3/2) and 726.5 eV (Fe 2p1/2) belong to Fe3+.At the same time,there are Fe satellite peaks at 718.5 and 732.1 eV[52].The results show that the Fe elements on the surface of each sample are composed of Fe (II) and Fe (III).

Fig.6.High resolution Fe 2p XPS spectra of (a) 5Fe-0,(b) 5Fe-25,(c) 5Fe-50 and (d) 5Fe-75.

The high-resolution C 1s XPS spectra(Fig.7)shows the peaks of four carbon species,which are C=C/C—C (283.6 eV),C—OH/C—N(285.1 eV),C—O—C (286.4 eV) and C=O/C=N (289.3 eV),respectively [53,54].Their relative contents are shown in Table 2.It is worth noting that 5Fe-50 and 5Fe-75 have more C—OH/C—N and C=O/C=N functional groups than 5Fe-0 and 5Fe-25.According to literature reports,C—OH/C—N and C==O may enhance the activation rate of PMS by accelerating electron transfer [53,54].

Table 2 The relative contents of C species in the catalysts

Fig.7.High resolution C 1s XPS spectra of (a) 5Fe-0,(b) 5Fe-25,(c) 5Fe-50 and (d) 5Fe-75.

The high-resolution N 1s XPS spectra of the 5Fe-25,5Fe-50 and 5Fe-75 catalysts with different urea addition are shown in Fig.8.5Fe-25,5Fe-50 and 5Fe-75 mainly contain five types of N species:the peaks at 397.8,398.9,399.7,400.9 and 403.7 eV are attributed to pyridinic nitrogen,FeN4,pyrrolic nitrogen,graphitic nitrogen and nitrogen oxide,respectively[55].The relative contents of each nitrogen species are shown in Table 3.It is found that the relative contents of pyridinic nitrogen,FeN4and graphitic nitrogen increase with the increasing nitrogen doping amount,but when the urea addition is higher than 50 mg,the increasing extents of these nitrogen slow down.It has been reported that graphitic nitrogen can promote electron transfer in the catalyst,pyridinic nitrogen as a Lewis basic site can improve the adsorption of electrophilic compounds (such as phenolic compounds and oxidants) [56],and it is easy to combine with Fe to form FeN4structure [57,58],and nitrogen oxides have no significant effect on the catalytic performance of carbon supported catalysts[59].5Fe-50 and 5Fe-75 have higher concentrations of graphitic nitrogen,pyridinic nitrogen and FeN4than other catalysts,which can provide more active sites for PMS activation and promote efficient degradation of pollutants.

Table 3 The relative contents of N species in the catalysts

Fig.8.High resolution N 1s XPS spectra of (a) 5Fe-25,(b) 5Fe-50 and (c) 5Fe-75.

Based on the above studies on catalyst characterization,we infer that there are Fe nanoparticles and FeN4active sites in 5Fe-50 catalyst,which are directly related to the synthesis process of 5Fe-50 (Fig.9).As shown in the figure,in the impregnation stage,Fe complexed with urea and the N element in urea anchored Fe during this process,then the complex is adsorbed on CMK-3-N surface and ensure Fe element highly dispersed on the support surface.In addition,a part of Fe can be directly adsorbed on the negatively charged CMK-3-N surface through electrostatic interaction(Fig.S4).After calcination,the complexes formed in the above process are connected with the carrier carbon network and form the FeN4sites.Moreover,the Fe atoms directly interacting with CMK-3-N in the calcination process can prevent the migration and aggregation of Fe due to the C3N4formation after urea pyrolysis.From Fig.7 and Table 2,it can be observed that there exist the oxygen-containing functional groups (C—OH/C—N and C=O/C=N)in the 5Fe-xcatalysts,implying the oxygenated edge defects have been introduced [46].And as the urea content increases,the content of C—OH/C—N functional groups first increases and then decreases,reaching its maximum value (13.5%) at a urea content of 50 mg;meanwhile,the content of the C=O/C=N functional groups continues to increase with increasing urea content,but the content of these functional groups in the 5Fe-75 catalyst is only 0.2% higher than in 5Fe-50.From Fig.8 and Table 3,it can be obvious that the characteristic peak attributed to FeN4is detected in the N 1s XPS spectra of the 5Fe-xcatalysts,demonstrating that the doping of Fe element promotes the generation of Fe based-N—C defects.By combining the results from HRTEM,it is found that the Fe nanoparticles supported on the surface of 5Fe-50 have no agglomeration and are highly dispersed,this suggests that the oxygenated edge defects over the carbon material and the carbon defect structure introduced by the heteroatom together promote the monodispersion of the active metal Fe on the catalyst surface,thus providing more active sites for the reaction.

Fig.9.Schematic diagram of the preparation mechanism of the 5Fe-50 catalyst.

3.2.Evaluation of catalyst performance in m-cresol degradation

3.2.1.Effect of N doping on catalytic performance

To investigate the influence of N doping amount on catalytic activity,the performance of 5Fe-0,5Fe-25,5Fe-50 and 5Fe-75 with various N doping levels to activate PMS form-cresol degradation was comparatively studied (Fig.10).As shown in the figure,when the urea dosage is less than 50 mg,them-cresol removal efficiency increases with increasing urea dosage;the 5Fe-50 catalyst has the bestm-cresol removal efficiency,reaching 100.0%m-cresol removal within 30 min,while the 5Fe-0 and 5Fe-25 catalysts only achieve 82.0% and 90.5%m-cresol removal within 30 min respectively.As the urea dosage is raised from 50 to 75 mg,the catalytic performance of 5Fe-75 does not improve considerably,and the reaction rate increases slightly within 20 min,reaching 100%mcresol elimination by 30 min.The performance of the catalyst is greatly increased with N doping,which is mostly because pyridinic nitrogen,pyrrolic nitrogen and graphitic nitrogen can activate PMS and thus generate ?OH and,and the graphitized nitrogen can promote electron transfer and carry more electrons to PMS to generate more radicals [60];in addition,the defective structures on the N-modified carbon carriers,such as edge defects,curvature and vacancies,can generate dangling σ-bonds,so that the πelectrons tend to be unrestricted by the edge carbon,and the electrons are donated from the carbon material to PMS,generatingand ?OH,as well aswith lower oxidation potential[55].Accordingly,50 mg of urea addition is chosen as the optimum urea addition to the catalyst in terms ofm-cresol removal efficiency as well as economics.

Fig.10.Effect of N doping amount on the degradation of m-cresol over the 5Fe-x catalysts (reaction conditions: m-cresol concentration=200 mg?L-1,catalyst dosage=0.4 g?L-1,PMS concentration=5 mmol?L-1).

3.2.2.Effect of reaction conditions on the catalytic performance of Fe and N co-doped ordered mesoporous carbon-based catalysts for the m-cresol degradation by PMS

3.2.2.1.Effect of PMS concentration on the m-cresol degradation.The concentration of PMS is an important factor affecting the reaction process.In general,the higher concentration of effective reactive oxygen species(ROS)generates after PMS activated,the more beneficial for the catalytic reaction proceeds.The effect of PMS concentration on them-cresol degradation by the 5Fe-50/PMS system is shown in Fig.11.From the figure,it can be seen that the degradation ofm-cresol firstly increases and then decreases with the increasing PMS concentration.As the PMS concentration increases from 1 to 5 mmol?L-1,the removal efficiency ofm-cresol increases from 71.0% to 100.0% within 30 min and reaches the highest catalytic efficiency at 5 mmol?L-1;as the PMS concentration increases to 7 mmol?L-1,the catalytic degradation efficiency decreases to 93% within 30 min.This result is attributed to the fact that with sufficient catalyst dosage,the PMS addition increasing in a certain range could produce more reactive oxygen species form-cresol oxidation;however,when the concentration of PMS is too high,thegenerated in the system will react with too muchbefore reacting withm-cresol (Eq.(2)) [33],andcan also undergo a self-quenching reaction to produce(Eq.(3))[61],thus causing the loss of the active species,which in turn leads to a decrease in the degradation efficiency ofm-cresol.According to the foregoing findings,the optimum PMS concentration is 5 mmol?L-1,and PMS concentrations that are either too high or too low would limitm-cresol degradation efficiency.

Fig.11.Effect of PMS concentration on the m-cresol degradation in 5Fe-50/PMS system (reaction conditions: m-cresol concentration=200 mg?L-1,catalyst dosage=0.4 g?L-1,PMS concentration=5 mmol?L-1).

3.2.2.2.Effect of m-cresol concentration on catalytic reaction efficiency.The impact ofm-cresol concentration on the degradation efficiency of 5Fe-50/PMS system is shown in Fig.12.As shown in the figure,the degradation rate and efficiency ofm-cresol are closely related to its initial concentration under the conditions of constant catalyst content (0.4 g?L-1) and PMS concentration(5 mmol?L-1),which are inhibited due to the increase of the initial concentration ofm-cresol.When them-cresol concentration is less than 200 mg?L-1,the degradation efficiency ofm-cresol reaches 100% within 30 min,and the efficiency reduces somewhat as the concentration ofm-cresol rises.When them-cresol concentration increases from 200 to 250 mg?L-1,the degradation efficiency ofm-cresol drops from 100% to 66% within 30 min.This could be because: at constant catalyst content and PMS concentration,the total amount of reactive oxygen species produced in the system is the same,but increasingm-cresol concentration accelerates the consumption of free radicals,lowering the probability of reactive oxygen species accessingm-cresol,as well as lowering the degradation rate and efficiency ofm-cresol [62].

Fig.12.Effects of m-cresol concentration on catalytic performance of 5Fe-50.

3.2.2.3.Effect of initial pH on the m-cresol degradation.Some studies have pointed out that the initial pH of the solution has an effect on the formation and conversion of sulfate and hydroxyl radicals in the system.Therefore,the impact of initial pH on the catalytic efficiency of the 5Fe-50/PMS system was systematically discussed.As shown in Fig.13,when the initial pH of them-cresol solution is not adjusted with acid-base regulators(pH=6.8),5Fe-50 shows excellent catalytic performance and can completely remove 200 mg?L-1ofm-cresol within 30 min;when the pH is reduced from 6.8 to 5.0,the degradation reaction rate increases;meanwhile,when the pH is elevated from 6.8 to 11.0,the degradation reaction rate and efficiency both decrease slightly.This phenomenon may be attributed to the different concentrations ofand ?OH generated at different pH.Studies have demonstrated that as pH increases,reacts with OH-and form ?OH,leading to a decrease inand an increase in ?OH[63],because the half-life ofis longer than that of ?OH,and hence the former is more likely to be in full contact withm-cresol,as pH increases,the oxidation capacity of ROS decreases slightly.Additionally,with the change of pH,the structure ofm-cresol present in the system is different: when the pH is acidic,m-cresol is present in the form ofm-cresol molecules,and with the increase of pH,m-cresol molecules are deprotonated and transformed intom-cresol anions[64,65].Under alkaline conditions,there is a repulsive effect between the 5Fe-50 surface andm-cresol anions due to their negative charges,which affects the catalyst’s ability to adsorb and degradem-cresol.The above study also confirms that the initial pH of them-cresol solution has no significant effect on phenol degradation,and the 5Fe-50/PMS system shows a wide range of pH adaptability.

Fig.13.Effect of initial pH on catalytic performance of 5Fe-50.

3.2.2.4.Effect of different anions and humic acid (HA) on the degradation of m-cresol.In the practical application of advanced oxidation technology based on sulfate radicals,the coexisting anions in water may affect the degradation efficiency of the SRAOP system by blocking the generation of reactive oxygen species or impeding the adsorption of pollutant substrates on the catalyst.Additionally,HA is a kind of natural organic macromolecular compounded with a basic structure consisting of aromatic and alicyclic rings,which are attached with oxygen-containing functional groups like hydroxyl,carboxyl and carbonyl groups;HA as a natural organic compound widely exists in water,but its influence on the activation of PMS to degrade pollutants is still unknown very clearly.So the research on the impact of coexisting anions and HA on the SR-AOP system is of great significance for the promotion of SR-AOP technology.The degradation ofm-cresol in Cl-(5 mmol?L-1),(5 mmol?L-1),and HA (5 mg?L-1) containing solutions at various concentrations was investigated,and the findings are given in Fig.14.The rate ofm-cresol degradation is significantly elevated in the presence of Cl-,which confirms that Clplays an important role in them-cresol/PMS system.This is likely due to the generation of more oxidizing reactive chlorine during the reaction (Eqs.(4)–(6)) [66],and the excess of reactive chlorine can also participate in the degradation ofm-cresol and promote the process of free radical reaction in conjunction with PMS [67].Meanwhile,has a less impact on them-cresol/PMS system,with just a minor increase inm-cresol degradation rate.Furthermore,the effect of HA on the degradation ofm-cresol shows that the catalytic performance is affected by the addition of 5 mg?L-1of HA into the system,with 93% and 98% ofm-cresol degradation within 30 and 60 min,respectively,andm-cresol not completely degrades within 60 min.This phenomenon may be attributed to:(1) humic acid molecules andm-cresol competitively consumeand ?OH in the system,i.e.,HA acts as a scavenger ofand ?OH,thus decreasing the degradation rate ofm-cresol [68];(2) functional groups on the surface of humic acid make it easier to be adsorbed on the catalyst surface,thus impeding the adsorption ofm-cresol and the activation of PMS by the catalyst[69].

Fig.14.Effects of different anions and humic acid on the removal of m-cresol.

3.2.2.5.Reusability of 5Fe-50 for m-cresol degradation.In order to obtain the repeatability of 5Fe-50,five times degradation reactions were carried out under the following conditions:m-cresol concentration=200 mg?L-1,the 5Fe-50 catalyst dosage=0.4 g?L-1,and PMS concentration=5 mmol?L-1(to recover the catalyst,the mixture after the reaction is filtered and washed three times with deionized water;it is then vacuum-dried at 80 °C for 2 h and reused;the lost catalyst is then replenished for the next cycle).As shown in Fig.15,the degradation efficiency ofm-cresol slightly decreases and remains at 90.5%within 30 min after five times catalytic cycling.The above results demonstrate that the 5Fe-50 catalyst has excellent cyclic stability.

Fig.15.Reusability of 5Fe-50 for m-cresol degradation (reaction conditions: mcresol concentration=200 mg?L-1,catalyst dosage=0.4 g?L-1,PMS concentration=5 mmol?L-1).

3.3.Quenching experiment and EPR test

To confirm the main reactive oxygen species formed in the 5Fe-50/PMS system,methanol,tert-butanol,p-benzoquinone and furfuryl alcohol were chosen as the free radical quenchers to carry out radical quenching experiments.Among them,methanol contains α-H in the structure,which has a good quenching effect on sulfate radicals and hydroxyl radicals [70];whereas there is no α-H in the structure oftert-butanol,so it can only quench hydroxyl radicals [71];and furfuryl alcohol can be used as a quencher for hydroxyl radicals and singlet oxygen [72];p-benzoquinone can only quench superoxide radicals (the reaction rate constants between the radicals and the quenchers are shown in Table S1)[73].As shown in Fig.16,when methanol,tert-butanol,pbenzoquinone and furfuryl alcohol are selected for quenching,the degradation efficiencies ofm-cresol within 60 min are 24%,44%,68% and 35%,respectively.According to the quenching results of methanol andtert-butanol,and ?OH are the main active species in 5Fe-50/PMS system,and the quenching results of benzoquinone also show thatcontributes to the degradation ofmcresol.Since furfuryl alcohol also has a strong quenching effect on ?OH,it is impossible to distinguish whether the substance quenched by furfuryl alcohol is ?OH or1O2.Therefore,the type of active species in the 5Fe-50/PMS system cannot be identified only by quenching experiment,and the specific active species that have been identified in the system should be further detected.

Fig.16.Radical quenching tests (reaction condition: MeOH concentration=TBA concentration=FFA concentration=200 mmol?L-1, p-BQ concentration=50 mmol?L-1, m-cresol concentration=200 mg?L-1,PMS concentration=5 mmol?L-1,catalyst (5Fe-50) dosage=0.4 g?L-1).

The specific reactive oxygen species in the 5Fe-50/PMS system are further verified by using EPR characterization.Because 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) can react withand?OH and form spin adducts of greater stability,it is used as a spin trapping agent forand ?OH in EPR measurement [74].It is clearly shown that there are obvious DMPO-?OH andsignals in the 5Fe-50/PMS system at 2 min,which indicates the formation of strong oxidizing substancesand ?OH[75,76].As the reaction continues,the strong signals of DMPO-?OH andcan still be observed at 6 min,and the peak signals are enhanced to some extent (Fig.17(a)).The characteristic signal ofindicates the formation of superoxide radical anion() (Fig.17(b)).Based on the previous studies by Qinet al.[77,78],is speculated to be generated by activating PMS with Fe (III) (Eq.(7)).Combined with the free radical quenching experiments,is also considered to play an important role in the degradation ofm-cresol in the 5Fe-50/PMS system.

Fig.17.(a) DMPO-?OH and DMPO-,(b) DMPO- and (c) TEMP-1O2 spin-tapping ESR spectra in various processes.

2,2,6,6-tetramethylpiperidine(TEMP)was used as a singlet oxygen (1O2) spin trapping agent to test whether there exists singlet oxygen in the 5Fe-50/PMS system (Fig.17(c)).It is obvious that no characteristic signal of TEMP-1O2is observed,indicating that only ?OH is quenched in the FFA quenching experiment.Moreover,the quenching effect of FFA is better than that of TBA,which is attributed to the larger reaction rate constant(1.5×1010)between FFA and OH than that(6.2×108)between TBA and ?OH.The above results show that the degradation ofm-cresol by 5Fe-50/PMS system is in a free radical pathway.

3.4.Catalytic reaction mechanism

According to the above analysis results,the existence of Fe species plays a significant role in the degradation ofm-cresol(Fig.S5),the reaction mechanism of the 5Fe-50 catalyst activating PMS form-cresol degradation could be deduced (Eqs.(8)–(14)).On the one hand,FeN4could induce a unique PMS activation process:the formation of highly dispersed FeN4coordination structure breaks the chemical inertness of sp2hybrid carbon,which absorbs electrons from surrounding carbon and induces the formation of positively charged carbon atoms,thus leading to the activation ofelectrostatically adsorbed on the catalyst surface and generating free radicals [79].This indicates that the formation of FeN4site in the catalyst ensures the active site of the catalyst no longer limited to the Fe active site,and part of carbon could also be used as the active site for PMS activation.With the same Fe addition,the N-doped catalyst exhibits more PMS activation sites and shows higher PMS activation performance.Moreover,the coordination structure of FeN4is uniformly dispersed on the surface of the catalyst,which maximizes the atomic utilization and significantly increases the activity of the catalyst.On the other hand,Fe(II)and Fe(III)in the coated Fe nanoparticles synergistically react withto generate:Fe(II)provides electrons to PMS to activate it and generate,which could continue to react with H2O and form the ?OH free radical.At the same time,Fe(III) reacts within solution and generates(it is speculated that the reaction rate of this reaction is small according to the effect ofquenching experiment),furthermore,Fe(III)can also react withto maintain the circulation of Fe(II) and Fe(III) (Fig.18) [80].The Fe and N co-doped 5Fe-50 catalyst constructs a special PMS activation system with the coexistence of Fe nanoparticle active site and highly dispersed FeN4site,and synergistic catalysis enables efficient catalytic activation of PMS at two types of Fe sites form-cresol degradation and promotes excellent catalytic performance.

Fig.18.Mechanism diagram of 5Fe-50 activating PMS.

4.Conclusions

A novel method for preparing the Fe-N co-doped carbon material with ordered mesoporous channels was proposedviaureaassisted impregnation and simple pyrolysis method.During the catalyst preparation,the complexation of N elements in urea could anchor Fe,and the formation of C3N4during urea pyrolysis could also prevent migration and aggregation of Fe species,which jointly improve the dispersion and stability of Fe.In the 5Fe-50 activated PMS system,the FeN4coordination structure could break the chemical inertness of sp2hybrid carbon,which absorbs electrons from surrounding carbon,induces the formation of positively charged carbon atoms and leads to the activation of;moreover,Fe(II)and Fe(III)in the supported Fe nanoparticles synergistically react withto generate.The co-operative catalytic effect of FeN4sites and highly dispersed Fe nanoparticles realize the dual-site PMS activation for highly efficientm-cresol degradation.Furthermore,the ordered mesoporous structure of the catalyst could improve the mass transfer rate of the catalytic process,which also promote PMS activation performance for the degradation ofm-cresol.This study can provide a theoretical guidance for the preparation of efficient and stable catalysts for the degradation of organic pollutants by activated PMS.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors gratefully acknowledge the financial support of the National Natural Science Foundation of China (22108145 and 21978143),the Shandong Province Natural Science Foundation(ZR2020QB189),State Key Laboratory of Heavy Oil Processing(SKLHOP202203008) and the Talent Foundation funded by Province and Ministry Co-construction Collaborative Innovation Center of Eco-chemical Engineering (STHGYX2201).

Supplementary Material

Supplementary data to this article can be found online at https://doi.org/10.1016/j.cjche.2023.06.026.